The extraction of image's salient points for image retrieval

A new salient point extraction method from Discrete Cosine Transformation (DCT) compressed domain for content-based image retrieval is proposed in this paper. Using a few significant DCT coefficients, we provide a robust self-adaptive salient point extraction algorithm, and based on salient points, we extract 13 rotation-, translation- and scale-invariant moments as the image shape features for retrieval. Our system reduces the amount of data to be processed and only needs to do partial entropy decoding arid partial de-qualification. Therefore, our proposed scheme can accelerate the work of image retrieval. The experimental results also demonstrate it improves performance both in retrieval efficiency and effectiveness.